Robust Wald-type tests based on minimum Rényi pseudodistance estimators for the multiple linear regression model

Impacto

Downloads

Downloads per month over past year

Castilla, E. and Martín, N. and Muñoz, S. and Pardo, L. (2020) Robust Wald-type tests based on minimum Rényi pseudodistance estimators for the multiple linear regression model. Journal of Statistical Computation and Simulation, 90 (14). pp. 2655-2680. ISSN 1563-5163

[thumbnail of castilla_robust.pdf] PDF
893kB

Official URL: https://doi.org/10.1080/00949655.2020.1787410



Abstract

We introduce a new family of Wald-type tests, based on minimum Rényi pseudodistance estimators, for testing general linear hypotheses and the variance of the residuals in the multiple regression model. The classical Wald test, based on the maximum likelihood estimator, can be seen as a particular case inside our family. Theoretical results, supported by an extensive simulation study, point out how some tests included in this family have a better behaviour, in the sense of robustness, than the Wald test. Finally, we provide a data-driven procedure for the choice of the optimal test given any data set.


Item Type:Article
Uncontrolled Keywords:Influence function; Minimum density power divergence estimator; Multiple regresion model; Rény pseudodistance; Robustness regression model; Rényi Pseudodistance; Robustness
Subjects:Sciences > Mathematics
Sciences > Statistics
ID Code:73978
Deposited On:21 Jul 2022 11:43
Last Modified:03 Aug 2022 09:06

Origin of downloads

Repository Staff Only: item control page